More Like this

1. BranchNet: A Convolutional Neural Network to Predict Hard-To-Predict Branches

   https://doi.org/10.1109/MICRO50266.2020.00022  

   Zangeneh, Siavash; Pruett, Stephen; Lym, Sangkug; Patt, Yale N. (October 2020, 53rd Annual IEEE/ACM International Symposium on Microarchitecture)

   null (Ed.)

   Abstract—The state-of-the-art branch predictor, TAGE, re- mains inefficient at identifying correlated branches deep in a noisy global branch history. We argue this inefficiency is a fundamental limitation of runtime branch prediction and not a coincidental artifact due to the design of TAGE. To further improve branch prediction, we need to relax the constraint of runtime only training and adopt more sophisticated prediction mechanisms. To this end, Tarsa et al. proposed using convo- lutional neural networks (CNNs) that are trained at compile- time to accurately predict branches that TAGE cannot. Given enough profiling coverage, CNNs learn input-independent branch correlations that can accurately predict branches when running a program with unseen inputs. We build on their work and introduce BranchNet, a CNN with a practical on-chip inference engine tailored to the needs of branch prediction. At runtime, BranchNet predicts a few hard-to-predict branches, while TAGE- SC-L predicts the remaining branches. This hybrid approach reduces the MPKI of SPEC2017 Integer benchmarks by 7.6% (and up to 15.7%) when compared to a very large (impractical) MTAGE-SC baseline, demonstrating a fundamental advantage in the prediction capabilities of BranchNet compared to TAGE- like predictors. We also propose a practical resource-constrained variant of BranchNet that improves the MPKI by 9.6% (and up to 17.7%) compared to a 64KB TAGE-SC-L without increasing the prediction latency. 

   more » « less
2. Integrating Flexible Normalization into Midlevel Representations of Deep Convolutional Neural Networks

   https://doi.org/10.1162/neco_a_01226  

   Giraldo, Luis Gonzalo; Schwartz, Odelia (November 2019, Neural Computation)

   Deep convolutional neural networks (CNNs) are becoming increasingly popular models to predict neural responses in visual cortex. However, contextual effects, which are prevalent in neural processing and in perception, are not explicitly handled by current CNNs, including those used for neural prediction. In primary visual cortex, neural responses are modulated by stimuli spatially surrounding the classical receptive field in rich ways. These effects have been modeled with divisive normalization approaches, including flexible models, where spatial normalization is recruited only to the degree that responses from center and surround locations are deemed statistically dependent. We propose a flexible normalization model applied to midlevel representations of deep CNNs as a tractable way to study contextual normalization mechanisms in midlevel cortical areas. This approach captures nontrivial spatial dependencies among midlevel features in CNNs, such as those present in textures and other visual stimuli, that arise from tiling high-order features geometrically. We expect that the proposed approach can make predictions about when spatial normalization might be recruited in midlevel cortical areas. We also expect this approach to be useful as part of the CNN tool kit, therefore going beyond more restrictive fixed forms of normalization. 

   more » « less
3. SE3Lig: SE(3)-equivariant CNNs for the reconstruction of cofactors and ligands in protein structures

   Bhadra-Lobo, Siddharth (December 2023, Proceedings of NeurIPS "Machine Learning in Structural Biology” workshop)

   Protein structure prediction algorithms such as AlphaFold2 and ESMFold have dramatically increased the availability of high-quality models of protein structures. Because these algorithms predict only the structure of the protein itself, there is a growing need for methods that can rapidly screen protein structures for ligands. Previous work on similar tasks has shown promise but is lacking scope in the classes of atoms predicted and can benefit from the recent architectural developments in convolutional neural networks (CNNs). In this work, we introduce SE3Lig, a model for semantic in-painting of small molecules in protein structures. Specifically, we report SE(3)-equivariant CNNs trained to predict the atomic densities of common classes of cofactors (hemes, flavins, etc.) and the water molecules and inorganic ions in their vicinity. While the models are trained on high-resolution crystal structures of enzymes, they perform well on structures predicted by AlphaFold2, which suggests that the algorithm correctly represents cofactor-binding cavities. 

   more » « less
4. LeanResNet: A Low-cost Yet Effective Convolutional Residual Networks

   Ephrath, J.; Ruthotto, L.; Haber, E.; Treister, E. (June 2019, Joint Workshop on On-Device Machine Learning & Compact Deep Neural Network Representations (ODML-CDNNR) at ICML 2019)

   Convolutional Neural Networks (CNNs) filter the input data using spatial convolution operators with compact stencils. Commonly, the convolution operators couple features from all channels, which leads to immense computational cost in the training of and prediction with CNNs. To improve the efficiency of CNNs, we introduce lean convolution operators that reduce the number of parameters and computational complexity, and can be used in a wide range of existing CNNs. Here, we exemplify their use in residual networks (ResNets), which have been very reliable for a few years now and analyzed intensively. In our experiments on three image classification problems, the proposed LeanResNet yields results that are comparable to other recently proposed reduced architectures using similar number of parameters. 

   more » « less
5. Real-time finger force prediction via parallel convolutional neural networks: a preliminary study

   https://doi.org/10.1109/EMBC44109.2020.9175390  

   Xu, Feng; Zheng, Yang; Hu, Xiaogang (July 2020, Proceedings of IEEE Engineering in Medicine and Biology Society Annual Meeting)

   null (Ed.)

   Continuous and accurate decoding of intended motions is critical for human-machine interactions. Here, we developed a novel approach for real-time continuous prediction of forces in individual fingers using parallel convolutional neural networks (CNNs). We extracted populational motor unit discharge frequency using CNNs organized in a parallel structure. The CNNs parameters were trained based on two features from high-density electromyogram (HD-EMG), namely temporal energy heatmaps and frequency spectrum maps. The populational motor unit discharge frequency was then used to continuously predict finger forces based on a linear regression model. The force prediction performance was compared with a motor unit decomposition method and the conventional EMG amplitude-based method. Our results showed that the correlation coefficient between the predicted and the recorded forces of the parallel CNN approach was on average 0.91, compared with an offline decomposition method of 0.89, an online decomposition method of 0.82, and the EMG amplitude method of 0.81. Additionally, the CNN based approach showed generalizable performance, with CNN trained on one finger applying to a different finger. The outcomes suggest that our CNN based algorithm can offer an accurate and efficient force decoding method for human-machine interactions. 

   more » « less